Liquid AI lifts the causal mask: decoders can be decision models too
helloiamleonie · x · 2026-11-06
Liquid AI's latest release combines the strengths of both encoder and decoder camps: they keep a decoder architecture but remove the causal mask.
The reasoning: a causal mask makes sense for models trained to predict the next token, but a model trained to make decisions should be allowed to attend to all tokens. Since the open ecosystem split into encoder-based and decoder-based approaches, Liquid — having both — tried each in their first iteration, and now opens up bidirectional attention for decision models.
Related event: Liquid AI Drops the Causal Mask to Turn Decoders into Decision Models(3 posts)→
More from Models
- Researcher: Opus over-refuses simple safety-topic questions, even on existing writeups — DimitrisPapail · 2026-11-06
- Apple's GSM-Symbolic paper: renaming numbers sinks SOTA models' math scores — imjustnewatai · 2026-11-06
- Oc-e-Non-v1: a tersified Qwen3-4B that cuts response length by 50%, built with Heretic — -p-e-w- · 2026-11-06
- Robot RSI: testing whether frontier models can set up robot environments — qinzytech · 2026-11-06
- NetEase Youdao's Confucius4-R2T2: Streaming ASR That Never Rewrites Its Output — dr_cintas · 2026-11-06
- Concept Country open-sources Qwen Concept 27B: 54% fewer thinking tokens via weight-space edits — GabGarrett · 2026-11-06